The following abbreviations may be found herein:
3GPPTHIRD GENERATION PARTNERSHIPPROJECTACKACKNOWLEDGEMENTBWBANDWIDTHCCECONTROL CHANNEL ELEMENTCRSCELL-SPECIFIC REFERENCE SIGNALDLDOWNLINKDM-RSDEMODULATION REFERENCE SIGNALeNB (OR eNodeB)LTE BASE STATIONHARQ (OR H-ARQ)HYBRID AUTOMATIC REPEAT REQUESTICICINTER-CELL INTERFERENCECOORDINATIONIEINFORMATION ELEMENTLTELONG TERM EVOLUTIONLTE-ALONG TERM EVOLUTION ADVANCEDMIBMASTER INFORMATION BLOCKNACKNEGATIVE ACKNOWLEDGEMENTOFDMORTHOGONAL FREQUENCY DIVISIONMULTIPLEXOFDMAORTHOGONAL FREQUENCY DIVISIONMULTIPLE ACCESSPBCHPHYSICAL BROADCAST CHANNELPCFICHPHYSICAL CONTROL FORMATINDICATOR CHANNELPDCCHPHYSICAL DOWNLINK CONTROLCHANNELPDSCHPHYSICAL DOWNLINK SHARED CHANNELPHICHPHYSICAL HARQ INDICATOR CHANNELPRBPHYSICAL RESOURCE BLOCKPUSCHPHYSICAL UPLINK SHARED CHANNELRERESOURCE ELEMENTREGRESOURCE ELEMENT GROUPTDDTIME DIVISION DUPLEXUEUSER EQUIPMENTULUPLINK
LTE wireless communication systems aim to provide enhanced services by means of higher data rates and lower latency with reduced cost. One benefit of deploying LTE TDD systems is to enable asymmetric UL-DL allocations in a radio frame. Typically if more data is to be sent in DL, there can be a higher number of DL subframes in a radio frame to accommodate that greater data volume. In LTE TDD systems, asymmetric resource allocation is realized by providing seven different semi-statically configured UL-DL subframe configurations for a given radio frame, as specified in Table 4.2-2 of 3GPP TS 36.211 v 10.5.0 (2012-06) which is extracted below.
TABLE 4.2-2Uplink-downlink configurationsDOWNLINK-UPLINK-TO-UPLINKDOWNLINKSWITCH-POINTSUBFRAME NUMBERCONFIGURATIONPERIODICITY012345678905 MSDSUUUDSUUU15 MSDSUUDDSUUD25 MSDSUDDDSUDD310 MS DSUUUDDDDD410 MS DSUUDDDDDD510 MS DSUDDDDDDD65 MSDSUUUDSUUD
In the table above, “D” indicates a DL subframe, “U” indicates an UL subframe and “S” indicates a special subframe. The different UL-DL configurations in the table provide between 40% and 90% DL subframes, and in conventional practice the UL-DL configuration in use at an eNB is informed to the UE (and changed) only via system information on the broadcast channel. The UL-DL configuration is only configured semi-statically and so may not adapt to the instantaneous traffic situation. This is inefficient in terms of resource utilization, particularly in small cells or cells with a small number of users where the traffic situation can often change rapidly.
To address this inefficiency, a flexible TDD configuration study item for LTE-A Release 11 (Rel. 11) was completed. Evaluations in the study item revealed possibly significant performance benefits by allowing TDD UL-DL reconfiguration based on traffic adaptation in small cells. The studies also recommend interference mitigation scheme(s) for systems with TDD UL-DL reconfiguration.
As with asymmetric UL-DL configuration and flexible TDD allocation, there are several challenges to overcome before any implementation may be considered viable. One particular challenge is to allow reconfiguration of TDD UL-DL configuration on at most a radio frame basis without significant impact on the current 3GPP specification, and to allow coexistence with legacy (i.e. Rel. 8, 9, 10) UEs. It is thought that improving PHICH resource allocation and maintaining backward compatibility may help in this regard.
FIG. 1 shows UE ACK/NACK procedure 100 (PUSCH transmission and PHICH reception) in TDD system. As an example of the LTE timing rules specified in Rel. 11 and illustrated in FIG. 1, one DL subframe may have the responsibility to send HARQ-ACK bits for 0, 1 or 2 UL subframe(s), depending on the adopted TDD configuration and DL subframe index. For instance:                If TDD configuration #0 is used, DL subframe #0 (101) in radio frame n is responsible for sending HARQ-ACK for two UL subframes (105), namely UL subframes #3 and #4 in radio frame ‘n−1’, and DL subframe #1 (102) in radio frame n is responsible for sending HARQ-ACK for one UL subframe, namely subframe #7 (106) in radio frame ‘n−1’.        If TDD configuration #2 is used, then DL subframe #0 (103) has no responsibility for sending HARQ-ACK feedback for any UL transmission and, as a result, there is no PHICH resource allocation on this DL subframe (103).        If TDD configuration #3 is used, then DL subframe #0 (104) is responsible for sending HARQ-ACK feedback for one UL subframe, namely subframe #4 (107).        etc        
The PHICH resource allocation on each DL/Special subframe for each TDD configuration is further illustrated in FIG. 2. The factor mi (as given by individual entries in the table) indicates how many copies of PHICH resource are assigned on a given DL subframe in the TDD system.
In Rel. 8, 9, 10 and 11, PDCCH is transmitted on CCE which is made up by REs (or REGs) not occupied by CRS, PCFICH and PHICH. The antenna port for CRS can be determined by blind decoding of PBCH and REs used for PCFICH are predefined. Although PHICH configuration can be determined from PHICH-Config via PBCH/MIB decoding, nevertheless for TDD systems it is still not sufficient to determine how many REs (or REGs) are used for PHICH. For example, for a configured DL system BW of 10 MHz, assuming Ng=1/6, then NPHICHGroup=2, it is still necessary to know the factor mi which depends on eNB configured UL-DL configuration in order to further determine the exact number of REs (or REGs) occupied by PHICH resource.
The value of mi is related to TDD configuration and DL subframe index. As a result, in order to figure out the REs used for PHICH and thus determine the REs (or REGs) that are carrying PDCCH(s), the TDD configuration should be determined first. However, TDD configuration is transmitted in SIB1 via PDCCH with an associated PDSCH and the problem becomes tricky because the index of TDD configuration is needed in order to decode PDCCH with an associated PDSCH which contains the index of TDD configuration. This is therefore a “chicken-egg” problem. One way to solve this problem, as was agreed in the standardization meeting, is that a UE should perform blind decoding by assuming on a particular subframe the possible value of mi in order to get the TDD configuration, and start to use the detected TDD configuration to decode other PDCCH.
FIG. 3 shows PDCCH blind decoding and PHICH resource assignment 300. As illustrated in FIG. 3, for different TDD configurations/DL subframes, the REG available for PDCCH transmission may be different, although PHICH configuration PHICH-Config stays the same                Case 1: No PHICH resource assignment, mi=0 4 REGs (301) are used for PCFICH transmission, all other REGs can be used for PDCCH transmission        Case 2: PHICH resource assignment for one UL subframe, mi=1 Beside the 4 REGs used for PCFICH transmission, another 6 REGs (302) are used for PHICH transmission, and the remaining REGs can be used for PDCCH transmission        Case 3: PHICH resource assignment for two UL subframes, mi=2 6 additional REGs are used for PHICH transmission, and the remaining REGs can be used for PDCCH transmission.        
From the above example, it can be appreciated that if a UE has an incorrect understanding of the mi value (which is determined by TDD configuration and subframe index), or PHICH resource is not assigned according to factor mi known by the UE, then the UE will have an incorrect understanding of CCE to REG mapping and will fail to decode PDCCH.